A method to denoise the epileptic EEG by EEMD and TFPF

In order to reduce the noise in the EEG signal and improve the signal-to-noise ratio, this paper proposes a new EEG filtering method based on the combination of ensemble empirical mode decomposition (EEMD) and time-frequency peak filtering (TFPF). First, the original signal is decomposed by EEMD to a series of intrinsic mode functions (Imfs) and the frequency components are from high to low. Then, for different Imfs, different window lengths are selected to filter them using TFPF, and the signal is reconstructed as the sum of the filtered and residual Imfs. Finally, sample entropy is extracted as the feature to classify whether the signal is epilepsy. Experiments show that this method effectively suppresses EEG noise and improves detection performance.